{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 63,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015873015873015872,
      "grad_norm": 8.639281115996736,
      "learning_rate": 2.6315789473684208e-08,
      "logits/chosen": -1.015625,
      "logits/rejected": -1.390625,
      "logps/chosen": -45.5,
      "logps/rejected": -80.5,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.15873015873015872,
      "grad_norm": 7.679101104209251,
      "learning_rate": 2.631578947368421e-07,
      "logits/chosen": -1.203125,
      "logits/rejected": -1.21875,
      "logps/chosen": -55.25,
      "logps/rejected": -57.5,
      "loss": 0.6937,
      "rewards/accuracies": 0.0694444477558136,
      "rewards/chosen": -1.6987323760986328e-06,
      "rewards/margins": -0.0013885498046875,
      "rewards/rejected": 0.00139617919921875,
      "step": 10
    },
    {
      "epoch": 0.31746031746031744,
      "grad_norm": 6.832273820892766,
      "learning_rate": 4.970588235294118e-07,
      "logits/chosen": -1.09375,
      "logits/rejected": -1.21875,
      "logps/chosen": -53.5,
      "logps/rejected": -56.75,
      "loss": 0.6895,
      "rewards/accuracies": 0.1875,
      "rewards/chosen": -0.01220703125,
      "rewards/margins": 0.0031280517578125,
      "rewards/rejected": -0.015380859375,
      "step": 20
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 7.271916625407268,
      "learning_rate": 4.676470588235294e-07,
      "logits/chosen": -1.1640625,
      "logits/rejected": -1.09375,
      "logps/chosen": -60.0,
      "logps/rejected": -56.75,
      "loss": 0.6819,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.04931640625,
      "rewards/margins": 0.033447265625,
      "rewards/rejected": -0.0830078125,
      "step": 30
    },
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 9.143992879642836,
      "learning_rate": 4.38235294117647e-07,
      "logits/chosen": -1.3203125,
      "logits/rejected": -1.2578125,
      "logps/chosen": -63.25,
      "logps/rejected": -56.75,
      "loss": 0.6607,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.01531982421875,
      "rewards/margins": 0.08056640625,
      "rewards/rejected": -0.095703125,
      "step": 40
    },
    {
      "epoch": 0.7936507936507936,
      "grad_norm": 8.656910281802439,
      "learning_rate": 4.0882352941176465e-07,
      "logits/chosen": -1.171875,
      "logits/rejected": -1.2109375,
      "logps/chosen": -64.5,
      "logps/rejected": -55.25,
      "loss": 0.6601,
      "rewards/accuracies": 0.38749998807907104,
      "rewards/chosen": 0.07958984375,
      "rewards/margins": 0.1142578125,
      "rewards/rejected": -0.03466796875,
      "step": 50
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 7.24400853291459,
      "learning_rate": 3.7941176470588235e-07,
      "logits/chosen": -1.25,
      "logits/rejected": -1.1953125,
      "logps/chosen": -62.0,
      "logps/rejected": -56.0,
      "loss": 0.6428,
      "rewards/accuracies": 0.375,
      "rewards/chosen": 0.099609375,
      "rewards/margins": 0.1171875,
      "rewards/rejected": -0.01708984375,
      "step": 60
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -1.25,
      "eval_logits/rejected": -1.171875,
      "eval_logps/chosen": -65.0,
      "eval_logps/rejected": -57.25,
      "eval_loss": 0.6325781345367432,
      "eval_rewards/accuracies": 0.4107142984867096,
      "eval_rewards/chosen": 0.06103515625,
      "eval_rewards/margins": 0.1689453125,
      "eval_rewards/rejected": -0.10791015625,
      "eval_runtime": 12.383,
      "eval_samples_per_second": 16.151,
      "eval_steps_per_second": 0.565,
      "step": 63
    }
  ],
  "logging_steps": 10,
  "max_steps": 189,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}